Thoughts on 1.05 Anvil of Krong?

Discussion in 'Dungeons of Dredmor General' started by rederick, Sep 10, 2011.

  1. rederick

    rederick Member

    Need more Ancient Kronian women!

    But no, really.

    Maybe I've just been immeasurably unlucky in my last run, but the success rate seems to have dropped somewhat, but that's not what I'm aiming towards here.

    Your item gets slammed, and slammed hard, on a failed enchant. I average +1 to something with a success, but every fail I've seen has been atleast -4 to something...s. I found my self circling ye olden anvil as I would an Ancient Kronian woman.

    Would she favor me with a kiss?
    Or would she stab me in the stomach with various weapons hidden about her person, and leave me on the floor cringing in agony?

    Found myself throwing junk gear on anvils. Stuff that had potential, just not enough for me to commit to. After a good item or two was utterly ruined by a curse and I had to kick it to the curb, I stuck to the junk gear enchanting. Then, inevitably, say after 3 successes, I'd get a curse, and that junk gear which was reaching the threshold to being equipped, had to be kicked to the curb.

    Ah, how fickle woman are. I mean, anvils. >_>

    Has anyone noticed this as well? Or have my curses been immensely unlucky? If it's a common trend, I'd almost say that 5th level of smithing would work well with an effect along the lines of "Your skill in smithing stays Krong's anger, reducing the potency of his curses when attemping to enchant an item." With a hard-minimum as -1 incase there's more fluxuation than I had seen with my trials.
     
  2. Haldurson

    Haldurson Member

    At least in 1.03, the number of blessings/curses that you get from krong was somewhat proportional to the number already there (base stats are not included in the calculation).
    /edit -- also your crafting skill was never affecting krong, which is why the text regarding crafting ability was removed in 1.05.
     
  3. IanExMachina

    IanExMachina Member

    I've honestly had really good runs with my anvils so far.
    Only had 1 curse over 6 floors so far.
     
  4. 123stw

    123stw Member

    @rederick
    You are right. The chance is still 5/7 to 2/7, but as you know the benefit varies between 1 - 3 (average of 2) while the penalty is consistently 4.

    so 2*5/7 - 4*2/7 = expected value of 2/7 positive per anvil.

    given you get 18 anvils (floor 10 omitted due to practicality), and half the stat being completely meaning to a specific build, you are looking at the average of 2.57 total meaningful positive stats for the whole game. Risk advert should avoid the anvils.

    So instead of stats, I burn anvils on crossbows for processing skills. Not shooting with them means I can dump any trash crossbow in and throw it away whenever it got cursed.

    @IanExMachina
    The whole nature of risk is that the max and min can vary a lot. So while you can get a lot of buff on this run, you can be getting cursed badly on another. If you enjoy the thrill of gambling that's fine, but don't expect it to have any consistency.
     
  5. rederick

    rederick Member

    So it is consistently 4 as a failed penalty. Then I'll stick to sub-par/dupe gear that I don't mind risking, that's just a bit too much of a gamble for my tastes. Thanks for the info!
     
  6. 123stw

    123stw Member

    @rederick
    Look at it this way, it really doesn't matter rather all the bonus are on 1 item or spread across. So why bother using it on good items to begin with?
     
  7. IanExMachina

    IanExMachina Member

    @123stw

    Oh no I was just providing a counterpoint to the unlucky rolls made by the OP.
     
  8. Kablooie

    Kablooie Member

    I haven't seen any issues with the Altar. But the Translation thing is irritating. I crafted six - that's SIX - Clockwork Bolt Throwers, and on every one of them, that "lost in the museum" message came up after only ONE or TWO attempts. I gave up trying to craft a decent CBT (but found an artifact one on 10th level, heh).
     
  9. pingchecker

    pingchecker Member

    I have always been a little bit confused about the anvil of krong. Lets say I find an artifact weapon and I place it on the alter, can that weapon be upgraded again? or does it just waste the anvil, and give you the possibility of worse stats.

    Basically what I am asking is, if you are very lucky can you keep dumping the same artifact weapon on the alter and get uber stat version of that unique weapon?
     
  10. 123stw

    123stw Member

    @pingchecker
    If you are extremely lucky (or manipulate RND with hard reset/read only), yes you can keep stacking positive stats on an item.
     
  11. Derakon

    Derakon Member

    Yeah, there's no Krong cap.

    What would be nice is if the screen you get when using a Krong item only showed the new bonuses/penalties on the item, making it easier to tell what's changed.
     
  12. Bhruic

    Bhruic Member

    The chances are correct, but the benefits are wrong. Positive benefits range from 1-2, with what I still consider a bug giving the possibility (6/82)^n of an extra "n" Primary Stats (in other words, the potential for an infinite number of Primary Stats is possible, but practically speaking, you'll rarely see an extra 1, let alone more than that). The negatives range from 2-4, with the same possibility of seeing extras.

    Excluding the "extras", the actual calculation would be more like:
    1.5 x (5 / 7) - 3 x (2 / 7) = 1.5 / 7

    Still a positive. But putting that into practise would only be feasible on large numbers of trials, the variability of small numbers would mean you'd get good and bad "runs" fairly commonly.
     
  13. 123stw

    123stw Member

    So to verify, positive starts with 1, then has a 7.3% chance of rolling 1 more stats, and continue to roll until it fails.

    Likewise, negative starts with 2, then has a 7.3% chance to roll 2 less stats, and continue to roll until it fails.

    But in that case 1.5 and 3 would not be the right number. Normal way of doing expected payoffs will be impossible because it will be infinite, but if you do a realistic cropoff (say 3 rolls), the result will only be 0.155 stat per anvil.
     
  14. Haldurson

    Haldurson Member

    Not true -- I've gotten at least +3 on several occasions, not including base stats. Very first Kronging in my last game gave me +2 to necro resist, and +1 to (I think) Toxic damage on a Toque of Kanada (it was the only actual usable loot I had found at that point in the game). I can't say how often it happens as compared with other bonuses, because I don't generally keep track (especially since my vision sucks and I can't always actually read the bonuses), but it does happen.
     
  15. 123stw

    123stw Member

    Look more closely, you do have about a 1/200 chance of getting 3 positive on a normal item. Also, anvil an item with artifact quality (such as Toque of Kanada), you also get an additional 1 time bonus.
     
  16. Bhruic

    Bhruic Member

    It is true, but it's also got exceptions, of which you've listed one. Items that are considered "artifact" class (check dredmorwiki.com for a list of what qualifies) have an "artifact quality" value, and that value gets used as the positive benefit number, rather than a randomly generated one. The Toque of Kanada is an "artifact", hence the discrepancy.

    No, what happens is that a it flips a coin (basically) and that's the number of bonuses it's going to apply. However, 7.3% of the time, it won't count the bonus that it selected towards that number. For example, if it's going to apply 2 bonuses, it might select +1 slashing damage, leaving it with 1 pick left. If it hits one of the 7.3% as its next pick, it won't decrease the number of picks left, leaving it still with 1 remaining. As long as it keeps hitting that 7.3%, it will continue to think it has 1 pick remaining, until it finally rolls something else, at which point it will have 0 picks left, and stop.
     
  17. 123stw

    123stw Member

    Okay..... So something like this?

    For positive

    Let Count = RNG between 1 - 2
    N = 0
    While N < Count
    If primary stat is roll (7.3%), + Bonus
    If all other bonus is roll (92.7%), + Bonus and + N

    For negative

    Let Count = RNG between 2 - 4
    N = 0
    While N < Count
    If primary stat is roll (7.3%), + Bonus
    If all other bonus is roll (92.7%), + Bonus and + N
     
  18. Bhruic

    Bhruic Member

    Mostly correct, only minor change is it would be:
    While N < Count

    -- And worth noting that the only 6 Bonuses that the 7.3% applies to are the Primary Stats
     
  19. 123stw

    123stw Member

    Yea I caught that mistake too.

    So it's not so much 7.3% of the time, more like if a primary stat is roll it doesn't add to N. I will go fix that.

    Ignoring the primary stat part, the expected payoff for 18 anvils is 3.86. Given at least half of them are useless for any particular build, the useful payoff is only 1.93.

    Also, what is the chances of getting a weapon process skill?
     
  20. Haldurson

    Haldurson Member

    What you are forgetting is that if a positive to a stat is truly useless, then their negative can't be all that bad either.

    Also is a negative from a base of zero harmful? For example, what's the difference between a blasting resist of zero and -1 (or -2, etc.). I'm not sure I know the answer, so I'm curious. If there is really no difference, then the negatives aren't as harmful.

    Furthermore, it's not always difficult to find (or craft) a replacement for a ruined item, except for the really top-tier stuff (which you can avoid kronging). For example, damaging a ring of iron thorns is, at best, a temporary setback if you are a smith, so those often are my first choice for krongings. Remember, that while there are random aspects, there are lots of choices involved that are NOT actually random. Add in Archeology, and you have even more options. Heck, towards the end game, where your weapons can get corrupted, you can always carry aroun decent spares (after all, all that other extra crafting crap you are carrying around that hasn't been converted to useful crap, can be lutefisked by level 10).

    So before you write off Krong, remember that
    1. With good choices, you can more often than not write off your negative results
    2. If many of the positive results aren't useful, that means that many of the actual negative results don't matter all that much either.